summaryrefslogtreecommitdiff
path: root/net
diff options
context:
space:
mode:
authorJunchang Wang <junchangwang@gmail.com>2010-12-08 16:55:16 +0000
committerDavid S. Miller <davem@davemloft.net>2010-12-10 15:36:52 -0800
commita8d764b9832d3cc86019f71916665dd2d337d7c2 (patch)
tree08637ac7b04a569b1b5152f8e52434c9f284bf29 /net
parent4afb7527ac8cc7bd8f03570e12f6eed0eca03363 (diff)
downloadlinux-3.10-a8d764b9832d3cc86019f71916665dd2d337d7c2.tar.gz
linux-3.10-a8d764b9832d3cc86019f71916665dd2d337d7c2.tar.bz2
linux-3.10-a8d764b9832d3cc86019f71916665dd2d337d7c2.zip
pktgen: adding prefetchw() call
We know for sure pktgen is going to write skb->data right after *_alloc_skb, causing unnecessary cache misses. Idea is to add a prefetchw() call to prefetch the first cache line indicated by skb->data. On systems with Adjacent Cache Line Prefetch, it's probably two cache lines are prefetched. With this patch, pktgen on Intel SR1625 server with two E5530 quad-core processors and a single ixgbe-based NIC went from 8.63Mpps to 9.03Mpps, with 4.6% improvement. Signed-off-by: Junchang Wang <junchangwang@gmail.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net')
-rw-r--r--net/core/pktgen.c2
1 files changed, 2 insertions, 0 deletions
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 2953b2abc97..18fe20dacc6 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2660,6 +2660,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
sprintf(pkt_dev->result, "No memory");
return NULL;
}
+ prefetchw(skb->data);
skb_reserve(skb, datalen);
@@ -3007,6 +3008,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
sprintf(pkt_dev->result, "No memory");
return NULL;
}
+ prefetchw(skb->data);
skb_reserve(skb, 16);